A novel control strategy for youBot arm

Nowadays robotics educational platforms play an important role in the development and testing of new robotic technologies. These systems offer at low cost the possibility to access and experiment basic robot functionalities. In this paper, we examined the KUKA youBot: an omnidirectional mobile platform equipped with a five DOFs (Degrees of Freedom) robotic arm. The current internal design of the control loops limits severely its performances. This paper aims to assess the existing controller limitations in order to re-design, develop and test a new controller for the youBot arm. This paper presents the architecture of the existing software, and identifies the bottlenecks of the control loops. After proposing a new interoperable architecture, we present a new hybrid motion/force decentralized control, which accounts for the system's dynamics. The new controller is evaluated by means of simulations and on the real youBot arm. The new controller solves the major limitations examined, enabling new control applications. Results show a better performance of the new controller, in particular the step response rise time of the joint position control is now 33% faster than using the old controller.